J . Am. Chem. SOC.1985, 107, 268-270
268
Table I. 13C Chemical Shifts of Cefoxitin (la) and Its Ammonolysis Products
la
carbon atom CZ CJ c 4 c6 c 7
cs
3-CH2 carbamate 4-CO2OCH, 7-NHCOR thienyl C H 2 thienyl C 2 thienyl C3
2a"
3a"
26.05 114.30 134.05 63.42 95.43 160.72 64.59 159.16 166.26 53.30 172.88 37.24 138.04
28.39 101.74 139.60 58.36 89.38 171.32 66.63 159.93 169.67 51.84 171.32 37.24 138.62
30.04 131.33 172.09 67.21 87.73 171.50 123.25 168.39 165.66 52.33 171.13 37.05 138.62
127.74, 127.91
not observed
127.55, 127.83
thienyl C4 thienvl .~C, r?
126.27 126.09 126.27 Numbering for 2a and 3a follows that of cefoxitin.
Table 11. I3C Chemical Shifts of Cephamycin C (lb) and Its Ammonolysis Products carbon atom
ca c0,-
lb -26.0 113.90 134.04 63.31 95.30 160.90 64.66 159.13 166.32 53.19 176.64 36.2 23.22 36.2 56.88 181.51
On the Antarafacial Stereochemistry of the Thermal [1,7]-Sigmatropic Hydrogen Shift
2b"
3b"
101.43 139.64 58.34 88.88 171.31 66.52 159.82 169.75 51.63 174.81 36.2 23.22 36.2 56.88 181.51
-30.0 131.31 172.37 67.29 87.34 171.49 123.23 168.37 165.76 52.30 174.81 36.2 23.22 36.2 56.88 181.51
Carl A. Hoeger and William H. Okamura* Department of Chemistry, University of California Riverside, California 92521 Received August 17, 1984 The classical thermal [ 1,714gmatropic hydrogen migration' is considered to be a pivotal event in the metabolic production of vitamin D.2 Although the intramolecular nature of this thermal process has been e ~ t a b l i s h e d ,and ~ , ~ the stereochemistry of the corresponding [ 1,5] migration has been demonstrated to be suprafacial,'v5 no direct evidence has yet been obtained for the antarafacial nature of the [ 1,7] process.' Our interest in the chemistry of vitamin D prompted us to prepare the labeled cisisotachysterols 1 and 2 and to study their thermal behavior.6 We herein wish to report the first example of the antarafacial nature of this rearrangement (Scheme I). The labeled cis-isotachysterols 1 and 2 were synthesized as outlined in Scheme I1 (steroid numbering). Treatment of Grundmann's ketone 36bwith trimethylsilyl iodide (generated in Scheme I
&:., C8H17
20
u
\H
C8H17
2b
"Numbering for 2b and 3b follows that of cephamycin C.
C8H17
-
2
2c
cephamycin C (lb) is treated similarly (Table 11). Analysis of the carbon-13 intensities of the methoxyl groups of species 1-3 affords approximate rates of P-lactam opening and side-chain expulsion. Although proton-decoupled carbon- 13 FTNMR spectra may exhibit systematic departures from quantitative accuracy due to differences in relaxation rates and Overhauser enhancement^,^ use of the methoxyl signals as measures of relative concentrations minimizes such factors. Intensity data for each cephamycin were analyzed by standard kinetics techniques6 and were found consistent with a reaction having two consecutive, first-order steps. The following rate constant values were calculated: kl, = 0.000 36 f 0.00004 s-I, klb = 0.000 18 f 0.00002 s-I, and kz. = kza = kl,. A weak effect of the 7-acyl group on the rate of (3-lactam cleavage was noted in that kl, was twice klb. In summary, we have characterized the intermediates of cephamycin ammonolysis and confirmed that such can occur via a two-step process. Although anhydrous ammonia is unlike physiological media, it does represent solvolysis in a dipolar, protic medium in which all species can be observed. This provides a unique means for gaining insight into (3-lactam chemistry not possible or recognized p r e v i o ~ s l y . ~ ~ ~
(6) Frost, A. A.; Pearson, R. G."Kinetics and Mechanism"; Wiley: New York, 1958; Chapter 8. (7) Hamilton-Miller, J. M. T.; Richards, E.; Abraham, E. P. Biochem. J . 1970, 116, 385. (8) Schanck, A.; Coene, B.; Dereppe, J. M.; Van Meerssche, M. Bull. SOC. Chim. Belg. 1983, 92, 81.
0002-7863/85/ 1507-0268$01.50/0
2;%
w
H
.I,.d
lo-d
2d
u
20-d
(I =
Suprafacial, i i = Antarofacial, H = Hydrogen Migration; D=Deuterium Migration)
(1) (a) Woodward, R. B.; Hoffmann, R. J. Am. Chem. SOC.1965, 87, 2511. (b) Spangler, C. W. Chem. Rev. 1976, 76, 187. (2) (a) Schlatmann, J . L. M. A.; Pot, J.; Havinga, E. Recl. Trau. Chim. Pays-Bas 196483, 1173. (b) Hollick, M. F.; Frommer, J.; NcNeill, S.; Richt, N.; Henley, J.; Potts, J . T., Jr. Biochem. Biophys. Res. Commun. 1977, 76, 107. (3) Akhtar, M.; Gibbons, G . J. Tetrahedron Lett. 1965, 509. (4) (a) Sheves, M.; Berman, E.; Mazur, Y.;Zaretskii, Z. J. Am. Chem. SOC.1979, 101, 1882. (b) Moriarty, R. M.; Paaren, H. E. Tetrahedron Lerr. 1980, 2389. ( 5 ) Roth, W. R.; Konig, J.; Stein, K. Chem. Ber. 1970, 203, 426. (6) (a) Hammond, M. L.; MouriBo, A,; Okamura, W. H. J . Am. Chem. SOC.1978, 100, 4907. (b) Condran, P. C., Jr.; Hammond, M. L.; Mourifio, A.; Okamura, W. H. Ibid. 1980, 102, 6259. (c) For references to related derivatives, see Onisko, B. L.; Schnoes, H. K.; DeLuca, H. F. J . Org. Chem. 1978, 43, 3441. Okamura, W. H. Arc. Chem. Res. 1983, 16, 81.
0 1985 American Chemical Society
J . A m . Chem. SOC., Vol. 107, No. I , 1985 269
Communications to the Editor Scheme I1 CEH17
hHI7
@+
.A.
(j3--H
2)PhSeCI, THF;
I) TMS-CI, L i I (TMSIpNH, CH2C12 (96%)
MCPBA, CH2CIp
I
n .
,
I
(85 % )
"ADA 7
'
Ho
W
qJy
8 ) n-BuLi, EtpO;
,
L
I M HOAC (86%)
w
3) NoBH4.
CF
5 ) PDC, PTFA, CHpC12 (91%) 6 )LiC e C H , THF (90%) 7) SOC12, C5HgN (67%)
lo*A;
-1
( 7 I %)
OTMS
9)tip/ Lindlor /C&, quinoline (67%) IO) NoBH4, CeC13, CH30H; HPLC
CeC13, CH3OH (94%)
6
w
vv I
(32%)
2
(53%)
+
W
Separation
O 2
Table I. 'H N M R Chemical Shifts (6) and Relative Integrations' compd
Hn 7'
HI
His
HI"
3.84 (0.92 f 0.01)[0.97 f 0.021 3.06 (0.99 h 6.22 (2.08 f 0.01)[2.03 f 0.021 5.52 (0.95 f 0.05)[0.98 f 0.031 Id (lS,lOR) 6.17, (2.05 f 0.04)[2.00 f 0.061 5.50 (0.88 f 0.03)[0.033 zk 0.0071 3.71 (0.95 f 0.04)[0.99 h 0.061 3.24 (1.02 f 0.071 2b (lR,lOS) 6.14 (2.08 f 0.02)[2.06 f ,011 5.49 (0.91 f 0.02)[0.89 f 0.031 3.70 (0.92 f 0.02)[0.94 h 0.011 3.20 (0.89 h f 0.0111 2d (IR,lOR) 6.25 (2.06 f 0.01)[2.08 f 0.031 5.49 (0.97 f 0.04)[0.022 f 0.0021 3.82 (0.94 f 0.01)[0.92 f 0.031 3.10 (0.99 h 0.071
I b (lS,lOS)
~
~~
0.06)c 0.05)[0.88 f 0.05)[0.053 0.04)[1.02 f ~~~~~
DThechemical shifts are for the unlabeled compounds, which a t 300 MHz (Nicolet 300, CDCI,) were identical with the observable resonances of the labeled series. The values given in parentheses are the relative integrated values for the unlabeled derivative; those given in brackets are for the corresponding labeled derivative obtained by heating 1 or 2. 'Chemical shift for H6,, is that for the center of the AB pattern. c N o signal was observed for the labeled material.
situ) in the presence of base' gave exclusively the thermodynamic trimethylsilyl enol ether 4. Reaction of the latter with benzeneselenenyl chloride followed by direct oxidation with MCPBA and thermolysis of the resulting selenoxide gave the enone 5.* Reduction of this enone with NaBH4 and catalytic deuteration of the resulting allylic alcohol gave the 14a,1Sa-dideuterio alcohol 7.9 Oxidation of this alcohol with pyridinium dichromate'O followed by addition of lithium acetylide to the ketone afforded a single propargyl alcohoPb which could be subsequently dehydrated with SOCll in pyridine to give the 15a-deuterio enyne 8 (>98% d , ) . The labeled enyne 8 could then be carried on through 9 to the cis-isotachysterols 1 and 2 via methods previously reported by this laboratory." It should be noted that the two-step sequence (7) (a) Miller, R. D.; McKean, D. R. Synthesis 1979, 730. (b) For a review concerning Me,SiI, see: Schmidt, A. H. Aldrichimica Acto 1981, 14, 31 and references therein. (c) Kruerke, U. Chem. Ber. 1962, 95, 174. (d) Schmidt, A. H.; Russ, M. Chem.-Ztg. 1978, 102, 26, 65. (8) (a) Reich, H. J. Acc. Chem. Res. 1979, 12, 22. (b) Reich, H. J.; Wollowitz, S . W.; Trend, J. E.; Chow, F.; Wendelborn, D. F. J . Org. Chem. 1978, 43, 1697. (c) Johnston, A. D. Ph.D. Thesis, University of California, Riverside, 1983. Early work on this conversion was also carried out by Dr. Antonio MouriRo. (9) In addition, a small amount of the 14&15,%dideuterio alcohol was formed (14a,15a:14&158 = 37:l). For recent related cases of hydroxyl-directed catalytic reductions, see the following: (a) Thompson, H. W.; McPherson, E. J . Am. Chem. Soc. 1974,96,6232. (b) Stork, G.; Kahne, D. E. J. Am. Chem. SOC.1983, 105, 1072. (c) Crabtree, R. H.; Davis, M. W. Organometallics 1983, 2, 681. (d) Corey, E. J.; Engler, T. A. Tetrahedron Lett. 1984, 26, 149. (e) Evans, D. A.; Morissey, M. M. J . Am. Chem. Soc. 1984, 106, 3866. (10) Corey, E. J.; Schmidt, G. Tetrahedron Lett. 1979, 399. (1 1) This is paper 28 in the series Studies on Vitamin D (Calciferol) and Its Analogues. For paper 27, see: Jeganathan, S.; Johnston, A. D.; Kuenzel, E. A.; Norman, A. W.; Okamura, W. H. J . Org. Chem. 1984, 49, 2152.
--
5 6 7 provides a simple solution to a classic problem in steroid synthesis, namely, the highly selective production of the transhydrindane nucleus. With the synthesis of the labeled cis-isotachysterols complete, their thermolytic behavior could be studied in a manner similar to that previously described for the unlabeled series.6b Scheme I shows the [ 1,7] hydrogen migration pathways possible for compounds 1 and 2. If the [ 1,7] migration is indeed an antarafacial process, one expects to obtain from the thermolysis of compound 1 only compounds l b and Id; conversely, suprafacial migration should give l a and IC. If a random process were operative, all four products (la-d) might be expected. Since the C, and C,, stereochemical assignements have been established earlier by independent synthesis of the corresponding unlabeled compounds, which in turn had been stereochemically correlated to 1a-hydroxycholesterol of known configuration,6bdifferentiation between the suprafacial and antarafacial modes is possible. A similar analysis of 2 reveals that 2b and 2d, and not 2a and 2c, should obtain from the antarafacial mode of rearrangement. The product mixture from heating a solution of 1 in isooctane ( M) at 98.4 "C (sealed tubes) for 26 h consisted of 22% 1, 47% l b , and 31% Id; that from heating 2 under identical conditions consisted of 22% 2, 15% 2b, and 63% 2d (HPLC analyses using calibrated columns). Preparative separation of the individual components was readily achieved by HPLC (Whatman M9 Partisil column, 10% EtOAc-hexanes). As summarized in Table I, the deuterium analyses of the products lb,d and 2b,d were conveniently carried out by 'H N M R integration of the resonances assigned to H1, Hlo, and H15of the products of 1 or 2. The observation of the products l b and Id from 1 and 2b and 2d from 2 thus demonstrates directly for the first time the antarafacial nature of the thermal [ 1,7] hydrogen migration. N
J. A m . Chem. SOC.1985, 107, 270-271
270
Acknowledgment. We are grateful to the National Institutes of Health (PHS Grant No. AM-16595) for financial support and to A. D. Johnston and A. MouriAo for several earlier contributions to this study. Dr. Menso Rappoldt of Duphar, B.V. (Weesp, the Netherlands), generously provided starting materials utilized in this research.
Scheme I
HO
0 5
Supplementary Material Available: IH NMR and other data for lb,d, Zb,d, 1-9, and unnumbered compounds ( 8 pages). Ordering information is given on any current masthead page.
/w OH
*OH
0
9
I
Concerning the Role of Nidurufin in Aflatoxin Biosynthesis
6
1
Craig A. Townsend*+ and Siegfried B. Christensen Department of Chemistry The Johns Hopkins University Baltimore, Maryland 21 21 8 Received June 4, 1984
’O
I
t
7 tl
OH
n
Experiments with specifically labeled specimens of averufin (1) have shown its intact incorporation into aflatoxin B, (Z).1-3 In
0
Ho
11
9
6‘
0 2
1
H
HO
0 0
OH
HO
0 3
4
particular racemic [ l’-2H,13C]averufin(1) was found to label (2-13 in 2 without detectable loss of deuterium relative to 13Cinternal standard, despite a net change in oxidation state at this carbon. This observation was interpreted4 as suggesting a pinacol-like rearrangement involving migration of the anthraquinone nucleus to (2-2’ in nidurufin (3) with departure of the 2’-hydroxyl whose exo orientation3v5 would particularly favor rearrangement on stereoelectronic grounds (Scheme I, path A). In this paper we examine the intermediacy of nidurufin (3) and its 2’-epimer pseudonidurufin (4) and, while both fail to give detectable levels ‘Research Fellow of the Alfred P. Sloan Foundation 1982-1986; Camille and Henry Dreyfus Teacher-Scholar 1983-1988. (1) Simpson, T. J.; deJesus, A. E.; Steyn, P. S.;Vleggaar, J. Chem. SOC., Chem. Commun. 1982, 631-632. (2) Townsend, C. A,; Christensen, S. B.; Davis, S . G. J. Am. Chem. SOC. 1982,104,6252-6153. Townsend, C. A.; Davis, S. G. J. Chem. Soc., Chem. Commun. 1982, 1420-1422. (3) The absolute configuration of averufin (1) is as shown and by inference (ORD/CD)5 that of nidurufin (3) (unpublished results, these laboratories in collaboration with Professor M. Koreeda, University of Michigan). The relative configurations of nidurufin (3) and pseudonidurufin (4) have been unambiguously established by total synthesis (Townsend, C. A.; Christensen, S . B., ref 4).5 (4) Townsend, C. A.; Christensen, S.B. Tetrahedron 1983,39,3575-3582. For a somewhat different proposal, see: Sankawa, Y.; Shimada, H.; Kobayashi, T.; Ebizuka, Y.; Yamamoto, Y.; Noguchi, H.; Seto, H. Heterocycles 1982, 19, 1053-1058. (5) The relative configuration of nidurufin was incorrectly assigned endo in the original isolation from A . nidulans: Aucamp, P. J.; Holzapfel, C. W. J . S. Afr. Chem. Inst. 1970, 23, 40-56. See also the isolation of 63-di-0methylnidurufin: Kingston, D. G. I.; Chen, P. N., Vercellotti, J. R. Phytochemistry 1976, 15, 1037-1039.
0002-7863/85/1507-0270.$01.50/0
8
of incorporation into aflatoxin B1 (2), further evidence is presented to support important limiting features of biogenetic Scheme IA. Racemic [ l’-2H]nidurufin (3) and [ l’-2H]pseudonidurufin (4) were ~ r e p a r e d by ~-~ extension of previously developed methods2V6 (both >98% d , ) . When administered to mycelial suspensions of Aspergillus parasiticus (ATCC 155 17) under conditions where averufin (1) gave >20% specific incorporation,2 3 and 4 gave no detectable incorporation into 2 (mass spectrum). Impermeability may be excluded as the mycelial pellets turn perceptibly from white to yellow with orange centers within 3 h of exposure to the labeled anthraquinones. However, unlike averufin (l), after an additional 3 h the medium gradually became deeper yellow in color as the administered 3 and 4 were excreted as polar, highly water-soluble conjugate^.^ This disappointing outcome was further tested using FLUFF, a variant of A . parasiticus isolated by Bennett,8 which produces a t most only a trace of aflatoxin and appears to be blocked before the anthraquinone stage of the pathway. Parallel experiments revealed that while added averufin (1) supported markedly enhanced aflatoxin production, the two hydroxylated derivatives 3 and 4 did not.9 Having now to exclude nidurufin (3) and pseudonidurufin (4) as efficient precursors of aflatoxin B, (Z), consideration of how the side chain of averufin itself might be transformed into the bisfuran leads to a mechanistic distinction that is accessible to experimental test (Scheme I). Path A, invoking oxidation at (2-2’ in averufin but not hydroxylation to nidurufin, would generate a reactive intermediate which itself may rearrange to 6,collapse to 7,and finally, upon Baeyer-Villiger-like reaction, yield versiconal acetate @).lo If in averufin (5) the 1’-oxygen were labeled (6) Townsend, C. A,; Davis, S. G.; Christensen, S . B.; Link, J. C.; Lewis, C. P. J . Am. Chem. SOC.1981, 103, 6885-6888. Townsend, C. A,; Bloom, L. M. Tetrahedron Lett. 1981, 3923-3924. (7) 48 h after administration of the anthraquinones, the medium had become bright yellow in the case of nidurufin and somewhat less intensely yellow for pseudonidurufin. Neither pigment could be extracted into organic solvent, but on standing in aqueous solution over 3 weeks, the conjugates had largely decomposed to cleanly return the respective labeled anthraquinones. (8) Bennett, J. W.; Silverstein, R. B.; Kruger, S. J. J. Am. Oil Chem. Soc. 1981, 58, 952A-955A. (9) The failure of the “diol” oxidation state to function in the masked pinacol rearrangement is reminiscent of the nonconversion of ent-6a,7a-dihydroxykaurenoic acid to GA,,-aldehyde in gibberellin biosynthesis: Hanson, J. R.; Hawker, J.; White, A. F. J. Chem. SOC.,Perkin Trans. 1 1972, 1892-1895. Graebe, J. E.; Hedden, P.; MacMillan, J. J. Chem. SOC.,Chem. Commun. 1975, 161-162.
0 1985 American Chemical Society
